Milos

/

slovak-gpt-j-405M

like 2

Hi, could we add some info how to fine-tune this model? Whatever I try with lora, it always fails. Probably because of mismatch of my training dataset with the pre-training datset.